Biomedical ontologies are increasingly important in genomic and proteomic research where complex data in disparate resources need to be integrated. In particular, the Gene Ontology (GO) has become a standard for genome annotation and the Open Biomedical Ontologies (OBO) is an umbrella for ontologies shared across biomedical domains. There is, however, a gap in the current OBO library-an ontology of protein classes and their relationships. The goal of the proposed project is to develop a PRotein Ontology (PRO) to facilitate protein annotation and functional discovery. PRO will be developed within the OBO Foundry, adopting principles specifying best practices in ontology development. The specific aims of this project are to: (i) develop a Protein Evolution (ProEvo) ontology for the description of proteins based on evolutionary relationships, (ii) develop an ontology for Protein Modified Forms (ProMod) for the representation of multiple protein forms of a gene, (iii) specify relationships between PRO, GO and other OBO ontologies, and (iv) disseminate PRO ontology and develop scientific case studies. ProEvo will be developed based on manually-curated families of full-length proteins in PIRSF and PANTHER and their constituent domains in SCOP and Pfam, initially for human protein-containing classes. The ProEvo classes and their relationships with GO terms will formalize the relationships between phytogeny and function, to allow more consistent and accurate inference of function based on experimental evidence. ProMod will define protein products generated by genetic variation, alternative splicing, proteolytic cleavage, and post-translational modification, initially for human and mouse proteins using fully-curated entries in UniProtKB/Swiss-Prot and the Mouse Genome Initiative, to support specific annotation of proteomes at the precise levels of variants, isoforms, and modified products. Defined based on scientific case studies of human and mouse disease proteins, the relations between PRO, GO, and other OBO ontologies, such as Disease Ontology, will capture the relationships required for disease understanding. For PRO .dissemination to the community, the ontology will be integrated into OBO, new relations will be added to the OBO Relations Ontology, and an annual protein ontology workshop will be organized. PRO will also be accessible from the PIR web site for integrative protein analysis. Through scientific meetings and collaborations, the PRO consortium will interact with the wider scientific community to ensure that PRO is useful and widely adopted. The PRO ontology will allow researchers to explore functional and evolutionary relationships of proteins to improve understanding of disease and identify potential diagnostic and therapeutic targets.